•62 REPOET OF THE 



higher members of the series abut upon the shore in a cliff 

 about seventy-five feet high. At the base we find four feet of 

 brown calcareous sandstone which is assumed to be next in 

 order above the oolitic beds of Mackinac Island. From this 

 point, the outcrop of the formation is traced in a ridge passing 

 between Grand and Long Lakes, in Presque Isle county, and 

 abutting upon the shore again at a point nearly opposite Mid- 

 dle Island. This island is made up of fragments of the lime- 

 stone. Gradually subsiding toward the south, the formation at 

 Thunder Bay Island rises barely to the surface of the water. 

 On the east side of the island, in the vicinity of the light-house, 

 it is seen forming vertical cliffs beneath the surface of the 

 water. In calm weather, upon a sunny day, the view of these 

 subaqueous precipices is truly impressive. Dark gorges, gloomy 

 caverns and perpendicular walls are seen dimly lit by the dimin- 

 ishing light, until darkness cuts ofi the view, and the plummet 

 feels its way to the depth of ninety feet, amongst the shadows 

 of the ruins of an ancient ocean stream: Passing hence under 

 the bed of the lake, the formation emerges on the Canadian 

 shore, between Douglass Point and Benson's Creek. It passes 

 thence in a broad belt to the shore of Lake Erie, which it occu- 

 pies between Buffalo and Long Point. Dipping toward the south- 

 west beneath a trough of newer rocks, it appears again upon 

 the northern shore of the lake between Point aux Pins and the 

 Detroit river, and passing into south-eastern Michigan, it arches 

 over, forming the anticlinal axis whose denudation has uncov- 

 ered the Onondaga Salt Group. From this axis it dips north, 

 south-east and south-west, passing beneath three distinct coal 

 basins. 



At the exposures of this group of rocks in the southern part 

 of the State, we find its thickness very considerably diminished. 

 The conglomerate, supposed to represent the Oriskany sand- 

 stone, has not been recognized. The thick brecciated mass is 

 not distinctly identifiable, though at Pt. aux Peaux and Stony 

 Pt, the formation is much broken up. Still the palseontological 

 -characters of the rock seem rather to ally it with that part of 



